#MARKETS & INVESTING
Share buybacks delivered 12.1% in abnormal returns over four years in the founding 1995 study. Correct for correlation across firms and the effect becomes undetectable.
Published on August 27, 2026
In 1995, David Ikenberry, Josef Lakonishok and Theo Vermaelen measured, in the Journal of Financial Economics, an average abnormal return of 12.1% over the four years following an open-market share repurchase announcement, and 45.3% for 'value' stocks. They read it as market underreaction. In 2000, Mark Mitchell and Erik Stafford revisited three samples of corporate decisions — acquisitions, equity issues and repurchases — in the Journal of Business. The standard method, they showed, assumes multi-year abnormal returns are independent across firms, when those firms often announce at the same time. Corrected for that correlation, the test statistics turn out to be up to four times too large.
Does a share buyback create value for shareholders?
The reference study measures a 12.1% abnormal return over four years. A methodological correction published five years later pushes that performance below the significance threshold.
In other words: the number exists; the demonstration much less so.
Four caveats, because they matter
- Both studies cover the US market over largely overlapping periods: the disagreement is not about the data but about how to test it.
- Mitchell and Stafford do not show the effect is zero, only that it can no longer be distinguished from chance once cross-correlations are corrected.
- The debate concerns long-run performance after the announcement, not the immediate price reaction.
- The accounting mechanics of a buyback are not in question: what is contested is the abnormal return attributed to it.
Why it still matters: a number quoted for thirty years may have been produced as much by a statistical assumption as by the market.
For you
- How many market 'facts' rest on an independence assumption that was never checked?
- Did you know the 12.1% abnormal return attributed to buybacks does not survive a correction for correlation across firms?
Source: Ikenberry, Lakonishok and Vermaelen, 'Market Underreaction to Open Market Share Repurchases', Journal of Financial Economics, 1995; Mitchell and Stafford, 'Managerial Decisions and Long-Term Stock Price Performance', The Journal of Business, 2000.
View the source →